Packing a chromatography column well comes down to creating a uniform, bubble-free bed of adsorbent material. The slurry method is the most reliable approach for most lab settings, and the entire process takes about 15 to 20 minutes once you have your materials ready. Here’s how to do it right, step by step.
Choose How Much Silica Gel You Need
The standard rule is to use 20 to 50 times the weight of your crude sample in silica gel (or alumina, depending on your separation). If you’re purifying 100 mg of crude material, you’d weigh out 2 to 5 grams of adsorbent. Use the higher end of that range for difficult separations where your compounds have similar polarities, and the lower end when the compounds in your mixture are well separated on TLC.
The column diameter matters too. You want enough silica to fill the column to a reasonable bed height without being so packed that flow becomes painfully slow. For most teaching lab and small-scale research columns, the bed height ends up somewhere between 10 and 20 cm.
Prepare the Column Before Packing
Start by clamping your column vertically in a ring stand or support rack. Push a small wad of glass wool into the bottom of the column using a glass rod, pressing it down to sit snugly above the stopcock. This plug keeps the silica from washing out the bottom. On top of the glass wool, add a thin layer of clean sand, roughly 3 to 5 mm deep. This creates a flat, even base for the adsorbent bed to rest on.
Close the stopcock. You’re now ready to pack.
The Slurry Method, Step by Step
The slurry method is preferred over dry packing for one important reason: when solvent first contacts dry silica or alumina, the interaction releases heat. If this exothermic reaction happens inside your column, it can boil solvent, generate bubbles, and crack the adsorbent bed. By making the slurry in a separate flask, you let that heat dissipate before anything goes into the column.
Weigh your silica gel into an Erlenmeyer flask. In a fume hood, add your starting eluent (the least polar solvent in your system, often hexane or petroleum ether) and stir with a glass rod. You’ll feel the flask warm up. Keep swirling until the mixture is completely wet, any gas bubbles have escaped, and the consistency is thick but still pourable, like a thin milkshake.
While the slurry settles, fill the column partway with the same eluent, leaving room for the slurry. Open the stopcock and place a beaker underneath to catch the solvent that drains through.
Now, in one quick, continuous motion, swirl the flask and pour the slurry into the column through a wide-mouthed funnel. Speed matters here. If you pour in stages, you risk creating layers with different densities, which ruins separation. Immediately rinse any remaining silica out of the flask with more eluent and pour that into the column too. Use a Pasteur pipette to wash any stray silica off the inside walls of the column reservoir.
Removing Air Bubbles
Air bubbles trapped in the packed bed are the single most common cause of poor separations. They create channels where solvent flows faster than through the surrounding silica, meaning your compounds travel unevenly down the column and your bands smear together instead of separating cleanly. Bubbles can also cause the adsorbent to crack as it dries around them.
Immediately after pouring the slurry, tap the column firmly and repeatedly with your knuckles or a cork ring. You should see small bubbles rise to the surface of the solvent. Keep tapping from different angles until no more bubbles appear. Then apply gentle air pressure to the top of the column (a rubber bulb or even blowing gently through tubing works) to compress the bed. Let solvent drain until it sits about 1 cm above the top of the silica. Do not let the solvent level drop below the top of the bed at any point from here on. Exposing the silica to air reintroduces the exact bubble problem you just solved.
If bubbles form later during a run, you can sometimes fix the problem by gently stirring the top portion of the resin with a clean spatula to release trapped air, then re-settling the bed. For stubborn bubbles below the bottom frit or filter, inverting the column briefly and applying gentle pressure to the sealed top can push air out through the exit tip. Degassing your solvents under vacuum before use prevents most bubble issues from developing in the first place.
Add a Protective Sand Layer
Once the bed is compressed and the solvent level is about 1 cm above the silica surface, carefully add a second thin layer of clean sand on top of the adsorbent. Keep this layer under 5 mm. Its job is to protect the flat, even surface of the silica bed when you add your sample. Without it, pipetting solvent onto the column tends to dig a small crater in the silica, creating an uneven starting line for your separation.
Loading Your Sample
You have two options for getting your sample onto the column: liquid loading and dry loading. Each has trade-offs.
Liquid loading is faster. Dissolve your sample in as little of your starting eluent as possible and carefully pipette it onto the sand layer. Let it drain into the bed, then rinse the walls with small portions of eluent to push all of the sample onto the column. The risk with liquid loading is reduced resolution. If you dissolve the sample in a solvent that’s too strong (more polar than your starting eluent), it pushes compounds through the column before they have a chance to interact with the silica, causing bands to overlap.
Dry loading gives better separation. Dissolve the sample in a strong solvent, add loose silica gel to the solution, then evaporate the solvent on a rotary evaporator. You end up with a free-flowing powder where your compounds are already adsorbed onto silica. Pack this powder into a small cartridge or sprinkle it on top of the sand layer. This approach produces sharper bands with minimal peak broadening because the sample starts as a thin, even layer rather than a dissolved plug of liquid. The trade-off is the extra preparation time.
For routine separations where your compounds are well resolved on TLC, liquid loading works fine. For tight separations or larger-scale purifications, dry loading is worth the effort.
Running the Column
Once the sample is loaded and has drained into the top of the bed, gently add eluent to the reservoir and open the stopcock. For gravity-driven columns, the flow rate depends on the column diameter and bed height. For flash chromatography, you apply compressed air or nitrogen to the top of the column to push solvent through faster, which is what makes flash columns practical for separations that would take hours under gravity alone.
Collect fractions in numbered test tubes or vials and monitor them by TLC to see which fractions contain your target compound. A well-packed column produces tight, symmetrical bands that elute cleanly into a small number of fractions. If your bands are broad, streaky, or overlapping, the most likely culprits are air bubbles in the bed, an uneven packing job, or a sample that was loaded in too strong a solvent.
Common Mistakes to Avoid
- Letting the column run dry. If solvent drops below the top of the silica, air enters the bed and creates channels. You’ll need to repack.
- Pouring the slurry in stages. This creates visible horizontal lines in the bed where density changes, and compounds will spread at each boundary.
- Skipping the tapping step. Even a few small bubbles can ruin an otherwise good separation. Tap thoroughly.
- Using too strong a solvent to dissolve the sample. Start with the weakest solvent that fully dissolves your material. If you load in pure ethyl acetate onto a hexane-packed column, your first compound will streak badly.
- Overloading the column. If you’re using less than 20 times the sample weight in silica, the column simply doesn’t have enough surface area to separate your compounds. Bands will co-elute.